README.md · Ekber/nlp-journey

# nlp journey

> Your Journey to NLP Starts Here ! 
> I sincerely welcome you to pull requests!

***全面拥抱tensorflow2，代码全部修改为tensorflow2.0版本。***

## 一. 基础知识

* [基础知识](docs/basic.md)
* [工具教程](tutorials/)
* [实践笔记](docs/notes.md)
* [常见问题](docs/fq.md)
* [实现代码](nlp/)

## 二. 经典书目([`百度云`](https://pan.baidu.com/s/1sE_20nHCfej6f9yRaisz7Q) 提取码：b5qq)

* 算法的乐趣. [`原书地址`](http://www.ituring.com.cn/book/1605) 
* 概率图入门. [`原书地址`](https://stat.ethz.ch/~maathuis/papers/Handbook.pdf)
* Deep Learning.深度学习必读. [`原书地址`](https://www.deeplearningbook.org/)
* Neural Networks and Deep Learning. 入门必读. [`原书地址`](http://neuralnetworksanddeeplearning.com/)
* 复旦大学《神经网络与深度学习》邱锡鹏教授. [`原书地址`](https://nndl.github.io/)
* 斯坦福大学《语音与语言处理》第三版：NLP必读. [`原书地址`](http://web.stanford.edu/~jurafsky/slp3/ed3book.pdf)
* CS224d: Deep Learning for Natural Language Processing. [`课件地址`](http://cs224d.stanford.edu/)

## 三. 必读论文

### 01) 模型与优化

* LSTM(Long Short-term Memory). [`地址`](http://www.bioinf.jku.at/publications/older/2604.pdf)
* Sequence to Sequence Learning with Neural Networks. [`地址`](https://arxiv.org/pdf/1409.3215.pdf)
* Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation. [`地址`](https://arxiv.org/pdf/1406.1078.pdf)
* Dropout(Improving neural networks by preventing co-adaptation of feature detectors). [`地址`](https://arxiv.org/pdf/1207.0580.pdf)
* Residual Network(Deep Residual Learning for Image Recognition). [`地址`](https://arxiv.org/pdf/1512.03385.pdf)
* Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. [`地址`](https://arxiv.org/pdf/1502.03167.pdf)
* How transferable are features in deep neural networks. [`地址`](https://arxiv.org/pdf/1411.1792.pdf)
* A Critical Review of Recurrent Neural Networks for Sequence Learning. [`地址`](https://arxiv.org/pdf/1506.00019.pdf)
* Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks. [`地址`](https://arxiv.org/pdf/1810.09536.pdf)
* Distilling the Knowledge in a Neural Network. [`地址`](https://arxiv.org/pdf/1503.02531.pdf)

### 02) 综述论文

* An overview of gradient descent optimization algorithms. [`地址`](https://arxiv.org/pdf/1609.04747.pdf)
* Analysis Methods in Neural Language Processing: A Survey. [`地址`](https://arxiv.org/pdf/1812.08951.pdf)
* Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer. [`地址`](https://arxiv.org/pdf/1910.10683.pdf)

### 03) 文本增强

* EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks.[`地址`](https://arxiv.org/pdf/1901.11196.pdf)

### 04) 文本预训练

* A Neural Probabilistic Language Model. [`地址`](https://www.researchgate.net/publication/221618573_A_Neural_Probabilistic_Language_Model)
* word2vec Parameter Learning Explained. [`地址`](https://arxiv.org/pdf/1411.2738.pdf)
* Language Models are Unsupervised Multitask Learners. [`地址`](https://d4mucfpksywv.cloudfront.net/better-language-models/language-models.pdf)
* An Empirical Study of Smoothing Techniques for Language Modeling. [`地址`](https://dash.harvard.edu/bitstream/handle/1/25104739/tr-10-98.pdf?sequence=1)
* Efficient Estimation of Word Representations in Vector Space. [`地址`](https://arxiv.org/pdf/1301.3781.pdf)
* Distributed Representations of Sentences and Documents. [`地址`](https://arxiv.org/pdf/1405.4053.pdf)
* Enriching Word Vectors with Subword Information(FastText). [`地址`](https://arxiv.org/pdf/1607.04606.pdf). [`解读`](https://www.sohu.com/a/114464910_465975)
* GloVe: Global Vectors for Word Representation. [`官网`](https://nlp.stanford.edu/projects/glove/)
* ELMo (Deep contextualized word representations). [`地址`](https://arxiv.org/pdf/1802.05365.pdf)
* BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. [`地址`](https://arxiv.org/pdf/1810.04805.pdf)
* Pre-Training with Whole Word Masking for Chinese BERT. [`地址`](https://arxiv.org/pdf/1906.08101.pdf)
* XLNet: Generalized Autoregressive Pretraining for Language Understanding[`地址`](https://arxiv.org/pdf/1906.08237.pdf)

### 05) 文本分类

* Bag of Tricks for Efficient Text Classification (FastText). [`地址`](https://arxiv.org/pdf/1607.01759.pdf)
* A Sensitivity Analysis of (and Practitioners’ Guide to) Convolutional Neural Networks for Sentence Classification. [`地址`](https://arxiv.org/pdf/1510.03820.pdf)
* Convolutional Neural Networks for Sentence Classification. [`地址`](https://arxiv.org/pdf/1408.5882.pdf)
* Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification. [`地址`](http://www.aclweb.org/anthology/P16-2034)

### 06) 文本生成

* A Deep Ensemble Model with Slot Alignment for Sequence-to-Sequence Natural Language Generation. [`地址`](https://arxiv.org/pdf/1805.06553.pdf)
* SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient. [`地址`](https://arxiv.org/pdf/1609.05473.pdf)
* Generative Adversarial Text to Image Synthesis. [`地址`](https://arxiv.org/pdf/1605.05396.pdf)

### 07) 文本相似性

* Learning to Rank Short Text Pairs with Convolutional Deep Neural Networks. [`地址`](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.723.6492&rep=rep1&type=pdf)
* Learning Text Similarity with Siamese Recurrent Networks. [`地址`](https://www.aclweb.org/anthology/W16-1617)
* A Deep Architecture for Matching Short Texts. [`地址`](http://papers.nips.cc/paper/5019-a-deep-architecture-for-matching-short-texts.pdf)

### 08) 自动问答

* A Question-Focused Multi-Factor Attention Network for Question Answering. [`地址`](https://arxiv.org/pdf/1801.08290.pdf)
* The Design and Implementation of XiaoIce, an Empathetic Social Chatbot. [`地址`](https://arxiv.org/pdf/1812.08989.pdf)
* A Knowledge-Grounded Neural Conversation Model. [`地址`](https://arxiv.org/pdf/1702.01932.pdf)
* Neural Generative Question Answering. [`地址`](https://arxiv.org/pdf/1512.01337v1.pdf)
* Sequential Matching Network A New Architecture for Multi-turn Response Selection in Retrieval-Based Chatbots．[`地址`](https://arxiv.org/abs/1612.01627)
* Modeling Multi-turn Conversation with Deep Utterance Aggregation．[`地址`](https://arxiv.org/pdf/1806.09102.pdf)
* Multi-Turn Response Selection for Chatbots with Deep Attention Matching Network．[`地址`](https://www.aclweb.org/anthology/P18-1103)
* Deep Reinforcement Learning For Modeling Chit-Chat Dialog With Discrete Attributes. [`地址`](https://arxiv.org/pdf/1907.02848.pdf)

### 09) 机器翻译

* Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation. [`地址`](https://arxiv.org/pdf/1406.1078v3.pdf)
* Neural Machine Translation by Jointly Learning to Align and Translate. [`地址`](https://arxiv.org/pdf/1409.0473.pdf)
* Transformer (Attention Is All You Need). [`地址`](https://arxiv.org/pdf/1706.03762.pdf)
* Transformer-XL:Attentive Language Models Beyond a Fixed-Length Context. [`地址`](https://arxiv.org/pdf/1901.02860.pdf)

### 10) 自动摘要

* Get To The Point: Summarization with Pointer-Generator Networks. [`地址`](https://arxiv.org/pdf/1704.04368.pdf)
* Deep Recurrent Generative Decoder for Abstractive Text Summarization. [`地址`](https://aclweb.org/anthology/D17-1222)

### 11) 关系抽取

* Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks. [`地址`](https://www.aclweb.org/anthology/D15-1203)
* Neural Relation Extraction with Multi-lingual Attention. [`地址`](https://www.aclweb.org/anthology/P17-1004)
* FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation. [`地址`](https://aclweb.org/anthology/D18-1514)
* End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures. [`地址`](https://www.aclweb.org/anthology/P16-1105)

### 12) 推荐系统

* Deep Neural Networks for YouTube Recommendations. [`地址`](https://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/45530.pdf)
* Behavior Sequence Transformer for E-commerce Recommendation in Alibaba. [`地址`](https://arxiv.org/pdf/1905.06874.pdf)
* MV-DSSM:A Multi-View Deep Learning Approach for Cross Domain User Modeling in Recommendation Systems. [`地址`](https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/frp1159-songA.pdf)

## 四. 必读博文

* 如何学习自然语言处理（综合版）. [`地址`](https://mp.weixin.qq.com/s/lJYp4hUZVsp-Uj-5NqoaYQ)
* The Illustrated Transformer.[`地址`](https://jalammar.github.io/illustrated-transformer/)
* Attention-based-model. [`地址`](http://www.wildml.com/2016/01/attention-and-memory-in-deep-learning-and-nlp/)
* Modern Deep Learning Techniques Applied to Natural Language Processing. [`地址`](https://nlpoverview.com/)
* Bert解读. [`地址`](https://zhuanlan.zhihu.com/p/49271699)
* 难以置信！LSTM和GRU的解析从未如此清晰（动图+视频）。[`地址`](https://blog.csdn.net/dqcfkyqdxym3f8rb0/article/details/82922386)
* 深度学习中优化方法. [`地址`](https://blog.csdn.net/u012328159/article/details/80311892)
* 从语言模型到Seq2Seq：Transformer如戏，全靠Mask. [`地址`](https://spaces.ac.cn/archives/6933)
* Applying word2vec to Recommenders and Advertising. [`地址`](http://mccormickml.com/2018/06/15/applying-word2vec-to-recommenders-and-advertising/)

## 五. 相关优秀github项目

* transformers. [`地址`](https://github.com/huggingface/transformers)
* HanLP. [`地址`](https://github.com/hankcs/HanLP)

## 六. 相关优秀博客

* [52nlp](http://www.52nlp.cn/)
* [科学空间/信息时代](https://kexue.fm/category/Big-Data)
* [刘建平Pinard](https://www.cnblogs.com/pinard/)
* [零基础入门深度学习](https://www.zybuluo.com/hanbingtao/note/433855)

Ekber/nlp-journey

简介

发行版

贡献者

近期动态

Ekber/nlp-journey .gitee-modal { width: 500px !important; }

简介

发行版

贡献者

近期动态

搜索帮助

Ekber/nlp-journey